Let The Data Tell You When You’re Done Doing Research

By jameszol
Published on December 4, 2007

One of our newest clients is a phenomenal company. Strong sales, rapid growth, poised for greater growth and they have a superior leadership team on all fronts…amazingly, they lack one thing: analytical web presence.

We just started tracking results accurately for the first time for them 30 days ago. It typically takes a little time to find the sweet spot for conversions using pay per click advertising with any client. There is always a sweeter spot – and that is where professional PPC Management comes in.

This case is unique though. We have no reliable history, no reliable data to review and analyze, conversions were severely skewed due to the wrong conversion codes being placed on landing pages, etc. We assumed that we could assess the situation accurately and correct the course in 30 days or less – our track record indicated this should be fairly straight forward work and it shouldn’t be a problem.

Well, 30 days is up and we look like a bunch of buffoons. Here is what we are learning from this experience:

1. Consider the history of the account. If there is NOT any accurate data depicting the specific action you want somebody to take, then do NOT assume the data can be readily available in 30 days or less – no matter what!

2. Do not make corrective decisions based on a single event. Patiently wait for more events to occur so you can be sure that a singular action was not a fluke, a random outlier, etc.

3. Communicate more effectively with the client. Create diagrams and build expectations based off research and historical data.

4. Do not RUSH the data by saying “close enough”. Your outliers could skew the standard if you do not have enough data and the decisions you make from the “close enough” research can lead you way off course and into the waters of high cost conversions.

5. Consistently under promise and over deliver. We strive to do this but we simply needed to educate this particular customer and re-educate ourselves on the specifics of research and analytics. It is extremely important to gather a large enough sample before making any critical changes and the sample size should be x number of conversions per keyword/ad/campaign/etc. We should wait until that data comes in BEFORE making any key decisions or we could lose our competitive foothold.

6. Every client and case is unique. While the function might be the same, the form will vary tremendously. We were looking at this particular account through our functional eyes when we needed to understand and analyze the form. The function in this case is our process for turning any account into gold and the form is the research, data, audience, and a variety of other factors that easily differ from client to client.

7. Let the data tell you when you’re done doing the initial research. Research 101 – and we forgot about it. Never, never, never tell the data it is done – bad decisions are made whenever there is an incomplete picture. In a lot of cases, the assumption is not worth the risk. Remember, when you assume, you make an ass out of u and me.

We are taking corrective action now, we have set another time frame but we are giving ourselves a fairly large window so a fair amount of data will be gathered and we can finally get some real work done. It’s time for us to work our magic and turn this client’s account into a bigger ROI machine. We’re fortunate that this client is very accepting and they understand the value of making good decisions based on a specific amount of data.

Bookmark or Share It Here:


If you enjoyed this post, make sure you subscribe to my RSS feed!

Leave a Reply

About Semvironment

Learn more about Semvironment, one of the world’s top pay-per-click marketing management companies.

Press Center

Feature articles and noteworthy instances of Semvironment in the news.

Hall of Fame

Read our most talked about news stories and PPC management training guides.